ai-lambda/app.py at main · stop-right-there/ai-lambda · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
import sys
import warnings
from datetime import datetime as dt
from datetime import timedelta

import torch
from flask import Flask, request
from flask_restx import Api, Resource
from model import GradePredictor, PathPredictor
from utils import *

app = Flask(__name__)
api = Api(app)
import os

import model
from serverless_wsgi import handle_request

# 사용자 정의 클래스를 찾을 수 있는 곳으로 sys.modules을 업데이트합니다.
# model = PathPredictor()
# # model.load_state_dict(torch.load("./checkpoints/state_dict.pth", map_location=torch.device('cpu')))
# model.load_state_dict(torch.load("./checkpoints/path_20230501_07445130_40960.001_state_dict.pth", map_location=torch.device('cpu')))
# model.load_scaler_params(np.load("./checkpoints/path_20230501_07445130_40960.001_scaler_params.npy"))
path_model = PathPredictor()
path_model.load_state_dict(torch.load("./checkpoints/path_20230501_07445130_4096 0.001_state_dict.pth", map_location=torch.device('cpu')))
path_model.load_scaler_params(np.load("./checkpoints/path_20230501_07445130_4096 0.001_scaler_params.npy"))

grade_model = GradePredictor()
grade_model.load_state_dict(torch.load("./checkpoints/grade_20230508_13504530_4096 0.0001_state_dict.pth", map_location=torch.device('cpu')))
grade_model.load_scaler_params(np.load("./checkpoints/grade_20230508_13504530_4096 0.0001_scaler_params.npy"))

# model.load_state_dict(checkpoint['model_state_dict'])
# 자기소개서 요약 API
@api.route("/api")
class Hello(Resource):
    def get(self):
        print("연결테스트")
        return {"content": "성공"}, 200

@api.route("/api/predict")
class Predictor(Resource):
    def post(self):
        try:
            query_hour = float(request.json["query_hour"])
        except Exception:
            query_hour = float(6)
        try:
            data = request.json["historical_details"] # Somehow get Data
            observation_date = data[0]["observation_date"]
            observation_date = dt.strptime(observation_date, "%Y-%m-%dT%H:%M:%S.%fZ")
            prediction_date = observation_date + timedelta(hours=query_hour)
            bef_data = parseJSON(data[1])
            cur_data = parseJSON(data[0], bef_data[0][2])
        except Exception as exp:
            return {"error_msg": "데이터를 파싱하는 과정에서 오류가 발생했습니다. \n오류 메시지:"+str(exp)}, 400
        data = np.concatenate([bef_data, cur_data], axis=0)
        data = np.expand_dims(data, axis=0)
        try:
            target_hour = torch.full((data.shape[0],), query_hour)
            path_predict = path_model.predict(data, target_hour=target_hour)  # Data should provided in shape of (1, 2, 97)
            grade_predict = grade_model.predict(data, target_hour=target_hour)
            print("prediction complete! path:", path_predict, "grade:", grade_predict, "query_hour:", query_hour)
        except Exception as exp:
            return {"error_msg": "예측 과정에서 오류가 발생했습니다. \n오류 메시지:"+str(exp)}, 400
        lat = float(path_predict[0][0])
        lon = float(path_predict[0][1])
        grade = int(torch.argmax(grade_predict[0])) + 1

        return {
            "timezone":"GMT",
            "units": {
                "grade": {
                    1: "TD",
                    2: "TS",
                    3: "STS",
                    4: "TY"
                },
                "central_pressure": "hPa",
                "maximum_wind_speed": "knot"
            },
            "prediction_date": prediction_date.strftime("%Y-%m-%dT%H:%M:%S.%fZ"),
            "central_latitude": lat,
            "central_longitude": lon,
            "grade": grade,
            "central_pressure": None,
            "maximum_wind_speed": None
        }, 200

def handler(event, context):
    return handle_request(app, event, context)
# if __name__ == "__main__":
#     app.run(debug=False, host="0.0.0.0", port=8000)